Active cooling in lights.

koti

Enlightened
Joined
Jun 21, 2008
Messages
316
Location
Warsaw
I thought I'd start a thread where we can gather all the info, projects (both existing and in the design phase) on lights utilising active cooling. There is not much out there so far that I know of so if you know of something interesting fire away !
Heres a an interesting project in the design phase to start the thread off:
http://lux-rc.com/content/products/flashlights/FB1_fat_boy/insider_files
 
Last edited:

CelticCross74

Flashlight Enthusiast
Joined
Aug 30, 2014
Messages
4,012
Location
Fairfax Va
I always thought of a liquid cooled light that would be kind of like how PC modders use some thicker liquids to cool their overclocked towers...interesting
 

idleprocess

Flashaholic
Joined
Feb 29, 2004
Messages
7,197
Location
decamped
Active cooling takes power - and typically no small amount of it - to be effective. It's more likely to be seen on the largest flashlights because they have enough watt-hours in their batteries to power cooling without impacting runtime or ergonomics (you're generally not going to pocket a big spotlight or lantern).

Something I'm surprised that we haven't seen is more use of heat pipes to move heat away from the LED's to more distributed heat sinks since heat will only move so quickly through a solid and heat sinks use fluid phase change to move it more quickly.

That Lux-RC project certainly is ambitious - 150W in a small package from what looks like 3 cells will be challenging technically and to the user. One suspects that runtimes will be short and the amount of time that someone wants to hold a small object dissipating that much power also short.
 

thedoc007

Flashlight Enthusiast
Joined
Feb 16, 2013
Messages
3,632
Location
Michigan, USA
Something I'm surprised that we haven't seen is more use of heat pipes to move heat away from the LED's to more distributed heat sinks since heat will only move so quickly through a solid and heat sinks use fluid phase change to move it more quickly.

Moving heat away from the LED and electronics is not the main problem, though. The main problem is how to remove heat from the light in general. Plenty of lights, even small ones (Niwalker MM15vn is one example) can run indefinitely at very high output if you run water over it. The batteries would be the limitation, not heat. But the flashlight to air interface just does not allow the heat to be removed as fast as it is generated. I'm not sure heat pipes would help significantly with that.
 
Last edited:

GordoJones88

Flashlight Enthusiast
Joined
Nov 26, 2011
Messages
1,157
Location
Tennessee
I keep my light on the end of a long string.
I spin it around in a circle over my head.
It completely lights up the area all around me.
Obviously the air circulation keeps the light cool even on Turbo.
 
Last edited:

idleprocess

Flashaholic
Joined
Feb 29, 2004
Messages
7,197
Location
decamped
Moving heat away from the LED and electronics is not the main problem, though. The main problem is how to remove heat from the light in general. Plenty of lights, even small ones (Niwalker MM15vn is one example) can run indefinitely at very high output if you run water over it. The batteries would be the limitation, not heat. But the flashlight to air interface just does not allow the heat to be removed as fast as it is generated. I'm not sure heat pipes would help significantly with that.

My sense (as an enthusiast, not an engineer) is that moving the heat rapidly away from the source would improve the overall heatsinking ability. I say this because many if my lights get intensely hot near the head but remain closer to room temperature near the midsection. If heat were transported more than uniformly, more of the surface area could dissipate the heat. Peak temperatures might be a bit lower at the LEDs.

But then again, implementing such a system would be difficult in something like a handheld flashlight - especially the standard tubular style which has so little volume to spare and generally must screw together.
 
Last edited:

Up All Night

Enlightened
Joined
May 29, 2012
Messages
487
I keep my light on the end of a long string. I spin it around in a circle over my head. It lights up a wide area of light around me like it's daytime. Obviously the air circulation keeps the light cool even on Turbo.
....I do the same. If the torch I'm using has a crenellated bezel .....it whistles!!
 

thedoc007

Flashlight Enthusiast
Joined
Feb 16, 2013
Messages
3,632
Location
Michigan, USA
My sense (as an enthusiast, not an engineer) is that moving the heat rapidly away from the source would improve the overall heatsinking ability. I say this because many if my lights get intensely hot near the head but remain closer to room temperature near the midsection. If heat were transported more than uniformly, more of the surface area could dissipate the heat. Peak temperatures might be a bit lower at the LEDs.

Good point, I hadn't really thought of that.
 

Fireclaw18

Flashlight Enthusiast
Joined
Mar 16, 2011
Messages
2,408
I saw a computer that did away with active cooling. Instead it had a giant copper heatsink on the top. The heatsink was fibrous, like a giant kitchen scrubber, so it had a ton of surface area.

Now to make a flashlight with a compartment behind the LEDs for a copper kitchen scrubber and a fan to blow air through it...
 

martinaee

Flashlight Enthusiast
Joined
Sep 16, 2012
Messages
1,495
Location
Ohio
Off topic, but imagine an XM18 with MT-G2s.

I think a lot of redesign would have to be done for that and probably would need more power. The best thing they could do at this point in time would be put xm-l2 leds in it. I wonder how many of those they sell-- it already puts out 18k lumens doesn't it so it's not like a lot of people are complaining about it being dim :)

I think they should make an xm-l2 version and put a video out. I honestly haven't payed much attention to them and it would bring a lot of instant focus back on their brand. 25-30k lumens out of that beast? Yes please.
 

Bright+

Newly Enlightened
Joined
Dec 5, 2008
Messages
170
I think a lot of redesign would have to be done for that and probably would need more power. The best thing they could do at this point in time would be put xm-l2 leds in it. I wonder how many of those they sell-- it already puts out 18k lumens doesn't it so it's not like a lot of people are complaining about it being dim :)

I think they should make an xm-l2 version and put a video out. I honestly haven't payed much attention to them and it would bring a lot of instant focus back on their brand. 25-30k lumens out of that beast? Yes please.

$2,400? What's the point of that thing?
 

thedoc007

Flashlight Enthusiast
Joined
Feb 16, 2013
Messages
3,632
Location
Michigan, USA
I think a lot of redesign would have to be done for that and probably would need more power. The best thing they could do at this point in time would be put xm-l2 leds in it. I wonder how many of those they sell-- it already puts out 18k lumens doesn't it so it's not like a lot of people are complaining about it being dim :)

I think they should make an xm-l2 version and put a video out. I honestly haven't payed much attention to them and it would bring a lot of instant focus back on their brand. 25-30k lumens out of that beast? Yes please.

Currently it does 15k lumens. And most XM-L2 LEDs are only about 10% brighter than XM-L, the new U3 maybe 15-20%? So it really wouldn't be a large upgrade, in my opinion. No reason not to do it for future orders (other than getting rid of old stock) but unless they drive them much harder, thus generating a LOT more heat, it would still be well under 20k lumens.

MT-G2 would indeed require a total re-design. The individual reflectors are quite small, and would have to be re-done. You'd need to add more cells. And the biggest problem, of course, would be heat. A fan just wouldn't cut it.
 

idleprocess

Flashaholic
Joined
Feb 29, 2004
Messages
7,197
Location
decamped
$2,400? What's the point of that thing?

Sizzle - a bleeding-edge demonstration of the technology that gets people excited so they go ahead and buy the steak they actually need/were going to buy ... from you, rather than a competitor.

It's similar to why automakers sponsor race teams - something exciting to associate their names with, but they won't sell you a race car. But if they did, it would be like 4 Seven's XM-18 ... expensive, impractical, unwieldy, and poorly suited to most peoples' actual needs.
 

koti

Enlightened
Joined
Jun 21, 2008
Messages
316
Location
Warsaw
Purely theoreticaly - Say we made a flashlight host out of enriched diamond which would have thermal conductivity close to 100%. Say the LED and the driver and all host parts are all made of enriched diamond so there would be no loss of heat conductivity at any point in the flashlight. Theoreticaly, the second we turn the light on the host should instantly have the same temperature as the emitter itself and we would be holding a boiling hot piece of diamond in our hands (say 10 non regulated MT-G's running off of 10 high drain 26650's in a smallest possible host)
Obviously this is a dead end...some kind of active cooling has to be implemented in order to get the heat out of that host. Heat management in turbocharged cars is fairly straightforward...an intercooler (secondery radiator just for the turbo) filled with a coolant does the job well. The bigger the intercooler, the bigger the turbo can be used with higher boost (higher volume of realy high temperatures) The thing is (and thedoc007 is right here) that in a car there is room for a big surface area intercooler, a flashlight host just wont be able to get rid of the heat that way.
 

mattheww50

Flashlight Enthusiast
Joined
Jun 24, 2003
Messages
1,048
Location
SW Pennsylvania
Peltier devices have a place, they are not well suited for applications where you need to move a lot heat without expending a lot of energy. They tend to be horribly inefficent. If you are lucky, you might be able to achieve a Coefficent of Perfomance of 1. That means that evey Calorie of heat energy you move,you expend 1 Calorie. So in order to dissipate 10 watts, you will consume another 10 watts doing it. Now instead of having to get rid of 10 watts, you need to get rid of 20 watts. Let me put the Coefficent of Performance into perspective. A top of the line energy efficient air conditioner/heat pump can achieve a Coefficient of Performance of about 4. (Coefficient of Performance is essentially SEER/4.184)
 

Bright+

Newly Enlightened
Joined
Dec 5, 2008
Messages
170
Let's provoke some thought.
There is certainly some overlap between actual technical and realistic benefits versus sales pitch.

We've had LED flashlights for many years before LED general lighting products became a common place like CFLs. These household LED lights are rated using Lighting Facts labeling. Creative specs writing on LED flashlight ratings is still at the whim of manufacturers marketing department.

I think running LEDs very cool is a marketing thing. LED bulbs get too hot to touch and their surface run up to 150-200F range without issues. These lights will get much more continuous hours and total hours realistically. When you compare it.. 1,000 hour life for all intents and purposes should be more than adequate for 99.99% of flashlights.

A heat spreader spreads heat evenly into the sink. A heat sink buffers the heat. So, a simple water filler with non-expanding freeze protection (so it does not crack upon freezing) is quite effective.

There's 28 18650s in there. Using 9Wh ea, you'll see that there's about 250Wh or 900kJ of available energy.

If there's a liter of water filler at 21C and you allow it to rise to 51C or a rise of 30C, it can absorb 125.5kJ or 35Wh with no heat removal at all. There is heat dissipation in real design, so you can do more in the real world. This can be explained by why it takes some serious time for a lava lamp to warm up. Water holds more heat per weight than any common material. It will cool down to ambient while it's switched off.

Realistically, what's the chance it will see continuous use for one hour set to 15000 "lumen" setting?
 

koti

Enlightened
Joined
Jun 21, 2008
Messages
316
Location
Warsaw
If there's a liter of water filler at 21C and you allow it to rise to 51C or a rise of 30C, it can absorb 125.5kJ or 35Wh with no heat removal at all.

I like that thought. It would not be practical in small hosts but for a big search light it just might work.
 
Top